Means for the determination of the dielectric permittivity of materials, based on the measurement of the reflection of electrical signals from the end of a dual-link coaxial cable upon its contact with the substance to be investigated are known (Yoshihito Hayashi et al., 2005, Phys. Med. Biol. 50, 599-612). However, due to the strong attenuation of the SHF signals in the coaxial (dual-link) cable at comparable wavelengths and cable radius, at frequencies above 10 GHz, performing measurements of complex dielectric permittivity becomes a very difficult technical problem.
In inventor's certificate SU No. 1830491, “Method of Determination of the Dielectric Parameters of an Object,” the measurement of complex dielectric permittivity of an object is accomplished on the basis of the parameters of reflected radiation of a specimen placed in the measuring line. The method envisages the fabrication of a specimen of specified overall dimensions and form. Thus, the claims contain the phrases “ . . . the specimen (material to be investigated) is made in the form of an insert, completely filling the cross-section of the transmission line, the coefficient . . . is measured . . . .” This approach is not suitable for the intravital measurement of biological tissues in vivo; a method permitting processing of the parameters of the signal reflected from an object of arbitrary form and dimensions, reflected from the boundary of the waveguide probe and the object being measured is necessary in this case.
Also, in inventor's certificate SU No. 1817555, “Method of Determination of the Dielectric Parameters of Materials,” a method is implemented on the basis of the parameters of the wave in the probing channel “with the specimen to be investigated.” A parameter of the specimen such as “1—specified length of the specimen to be investigated” is contained in the distinctive part and in the calculation formulas.
Patent RU No. 2 078 336, “Method of Monitoring the Properties of Materials on the Basis of Dielectric Loss Factor Variation and a Device for Implementing It,” is also known, in which an alternative is proposed to the condenser method in the mid- to high-frequency region with the use of a mounted capacitance sensor in whose electrical field the material to be investigated is situated. A device implementing this method operates in “a wide range of frequencies (100 kHz-10 MHz) . . . ,” i.e., in the radio range, maximum up to tens of megahertz. However, the circuit design of this device does not envisage operation in the range of tens of GHz, for which the device being applied for is proposed.
A method (Patent RF No. 2098016, priority of 30.01.97) of measurement of the dielectric parameters of bodies at a frequency of 30 GHz by means of a reflection SHF-dielectrometer with a single-link (circular) waveguide probe is also known.
The method includes the generation of an SHF signal, its division into a reference and a measuring signal, irradiation of the object to be measured by the latter, reception of the reflected signal modulated by a low-frequency signal, and then the determination of the physical parameter of the irradiated object from the reflected and the resultant signals. At the same time, the circuits of the measuring and the reference signals are equipped with waveguide transformers that provide the possibility of fine-tuning the path length, that in the process of measurement shift the phase of the measuring signal by π, and of the reference signal by π/2, radians.
A deficiency of said method and of the device implementing it is the necessity of observing the condition of good emitter matching (the waveguide probe) during irradiation of a material with specified reflective properties and dielectric permittivity close to those expected for the materials being tested. The absence of good emitter matching of the dielectric properties of the emitter and the materials being tested at the least reduces measurement accuracy, while a substantial mismatch of the dielectric properties of the emitter and the material being measured even in principle will prevent measurement of the dielectric parameters. This is explained by the following. The existence, at the open end-face of any waveguide, of several waves (modes) with different wave numbers and phase velocities is inevitable. The presence of several (and even more so, many) waves with different phase velocities prevents the obtaining of a stable interference pattern as the reflected and reference signals are superimposed. Thus, a single-mode waveguide is indispensable for unambiguous interpretation of the measurement results. However, single-mode radiation in an infinite waveguide is transformed into multimode radiation at the end-face of an open waveguide. The redistribution of energy to higher modes will be quite significant, because at the dielectric-filled (to ensure matching with the measurement medium) terminal segment of the waveguide probe, some of the local damping modes of the empty waveguide, having an imaginary wave number, will become propagating modes, i.e., having a real number, and the dielectric waveguide that is matched with respect to one mode will nonetheless be left mismatched with respect to other modes. Modes with several numbers n and m differing from zero will be propagating modes in a dielectric-filled waveguide with permittivity “∈,” since the expression for the longitudinal wave number will be greater than zero (∈·k2−(mπ/b)2−(nπ/a)2>0) for several combinations of numbers n and m, whereas in the empty waveguide the condition (k2−(mπ/b)2−(nπ/a)2>0) is observed only for m=0 and n=1. Here ∈ is the relative dielectrical permittivity of the substance filling the waveguide probe; k is the wave number of the probe radiation in air (vacuum); a and b are the dimensions of the cross-section of the waveguide; and m and n are integers. In summary, multimode radiation emerges inevitably at the end-face of an open waveguide; however, after reflecting from the material being measured and passing further from the end-face to the detector in the empty waveguide duct, the reflected multimode radiation is converted nonetheless to single-mode. In a homogeneous waveguide, at a distance literally of one to two wavelengths from the boundary of the dielectric insert, only single-mode radiation is left as a result of the redistribution of the energy of the higher modes to the fundamental propagating mode. Nevertheless, local multimodality of the radiation, even only at the end-face, inevitably leads to a complex value of the impedance of the entire waveguide probe. A method of compensation of multimodality of the radiation at the end-face of a waveguide by means of waveguide transformers that achieve matching of impedances of a waveguide probe and the substance being measured is known (RF patent 2098016-prototype). In the device described in this patent, the multimodality of the reflected signal is not manifested only in the case of interaction with a load that is matched with the waveguide probe, but is not arbitrary. The obvious inconvenience of such a technical solution resides in the fact that, having achieved matching (equality) of the impedances of a waveguide and the substance being measured, it is further necessary to determine the impedances of the matched waveguide probe in the measuring line. In the end, measurement on such an instrument requires a set of standard specimens with known components of dielectric permittivity, and the measurement of the dielectric permittivity of an unknown material will correspond to the choice of a standard specimen that is maximally close by degree of matching (equality of the impedances securing the minimum SWR—standing-wave ratio). And the more accurate the measurement is required to be, the greater the number of standard specimens it will be necessary to have: the set of standard specimens must form a “matrix” of materials with fine gradation of their dielectric properties with respect to ∈′ and ∈″—the real and imaginary components of the complex dielectric permittivity. For example, if restricted to a “step” of unity with respect to ∈′ and ∈,″ to measure the moisture content of material with an accuracy no worse than 1.5% (for such accuracy, a unit “step” with respect to ∈′ and ∈″ will at a minimum be sufficient), for an instrument at a frequency of 30 GHz a set of no fewer than 35·40=1400 standard specimens will be required; this makes the practical application of such a instrument difficult and significantly increases the cost of its production. The objective of the proposed solution is to increase the accuracy of measurements by achieving the independence of the process of measurement from matter of the matching of the impedances of the waveguide probe and the material being investigated.